UC Berkeley Health Records Hacked

This article talks about how the medical records databases at UC Berkeley were hacked. This illustrates the cost/risk issue of central servers. On one hand, having central servers (highly connected nodes) make it easy to store information and get it to the right people. On the other hand, they run a higher risk of exposure because of the numerous edges, and more information is easily accessed by the wrong people through these servers.

http://www.pcworld.com/businesscenter/article/164640/hackers_break_into_university_health_records.html

It would be very interesting to see you could use the network structure to your advantage to find out if people are hacking in. For instance, if you model the traffic in the network as a traffic game like we have been doing in class, you could determine the best way to route users through the network. If you could monitor how many people we're using each edge, you could tell when there was an extra user, i.e. a hacker. This of course assumes that you can force people to take certain routes in the network, and know how many people should be there at a certain time. While it doesn't necessarily stop unauthorized people from accessing the network, it might make it easier to detect them.